It is undeniable that services for migrants are dynamic and impacted not only by providers’ individual attitudes, but also by the health need of migrants and their family members, as well as the influence of underlying health system, legal implication and social values. Regarding the review findings, two conceptual themes were identified, namely: ‘Complexities in managing health care in a culturally-sensitive manner in light of resource constraints and the fear of making cultural mistakes’, and ‘Professional ethics in light of restrictive healthcare policy’. Note that relevant references, which failed to pass the screening process and did not appear in the data extraction table, might also be discussed in the following section in order to support or contest the review findings.
Complexities in managing health care in a culturally-sensitive manner in light of resource constraints and the fear of making cultural mistakes
Theoretically, culturally-sensitive health care is perceived as an effective means for promoting better health statuses for immigrants [
62,
63]. This can help overcome the powerlessness that immigrants often feel when they are excluded from the dominant culture of their destination countries. However, applying the concept of cultural sensitivity into real practice is not straightforward. The problems of using an interpreting service from the review above provide an obvious example. Akhavan [
26] and Farley et al. [
35] suggested that while healthcare providers recognised the merit of using an interpreting service, the practice of using interpreters was labour intensive and time consuming. These challenges presented in a context in which healthcare providers already faced a shortage of financial and human resources. The congruent findings reported by Eklof et al. [
33] and Lindsay et al. [
44], demonstrated that using phone interpretation services significantly increased the workload of nursing staff. Bischoff and Hudelson [
64] provided additional evidence which showed that, although, using ‘professional interpreters’ was considered as ‘gold standard’ for providing multi-cultural care, hiring bilingual interpreters did not guarantee high quality culturally-sensitive care. Specifically, problems arose when the interpreters were able to overcome the ‘language’ difficulties but still lacked a clear understanding of migrants’ behaviours and beliefs [
64,
65]. Binder et al. [
66] and Lyberg et al. [
45] found that in some circumstances, such as, during delivery and maternity care, interpretation services were of little use. In summary, the complexity of overcoming language barriers could not be solved solely by ‘hiring’ interpreters; successfully overcoming these barriers also required identifying whether or not the service matched needs of migrant beneficiaries.
While many studies recommended the use of professional interpreters, there is still room for using family interpreters. Although the use of family interpreters was not considered standard clinical practice physicians still accepted this in some situations, specifically when clinical presentations were uncomplicated (e.g., cough, cold, fever, etc.). Gray et al. [
67] suggested that refugees and migrants with limited English proficiency (LEP) in New Zealand preferred relying on their bilingual relatives to using professional interpreters. In contrast to Australia, where telephone interpreter was freely accessible, healthcare providers in New Zealand had to shoulder the cost of a landline interpreting service; this resulted in a low utilisation rate of telephone interpretation support [
67]. Some providers avoided face-to-face interpretation by preparing translated materials, like leaflets or videotapes, which were especially useful in maternal care. Yet, the efficacy and effectiveness of using those materials had not been explored [
45].
Another noteworthy issue was identifying how to deliver healthcare services in a ‘culturally-sensitive’ manner without creating a sense of ‘discrimination’ or ‘racism’. An obvious instance was depicted by Manirankunda et al. [
46] in Flanders, Belgium. The study was centred on provider-initiated HIV testing and counselling (PITC) for Sub-Saharan African migrants (SAMs). Though PITC was deliberately initiated to tackle high prevalence of HIV/AIDS amongst the SAMs population, physicians were reluctant to encourage their patients to undertake HIV/AIDS testing (unless patients themselves requested) owing to a fear of being accused of racism. A survey of the opinions of health experts in Greece failed to reach a consensus on whether or not establishing a separate ward for migrants would increase the efficiency of healthcare service delivery, this resulted from fears that it might emphasise the perception of racial inequity [
68]. This is something Worth et al. [
61] called the ‘fear of making cultural blunder’ in health practitioners. In occupational health, even though racism is not expressed explicitly, Meershoek et al. [
69] observed that Dutch doctors assigned a stereotype of ‘problematic’ patients to migrants more often than to Dutch patients. This bias explained why Dutch occupational physicians occasionally failed to fine tune their coaching activities to meet the needs of migrants, and this, in turn, made some migrants more likely to suffer from conditions that prevented them from working than general Dutch patients.
A similar issue was found in debates on tuberculosis screening of asylum seekers in the UK; is it a protective measure for the benefit of all UK residents, or another kind of racial discrimination [
70]? Bracanovic [
71] argued that being ‘culturally sensitive’ in bioethics was implausible for the following reasons: (1) it rendered the disciplinary boundaries too flexible and was inconsistent with Western biomedical sciences, (2) it was practically useless because it approached cultural phenomena in a predominantly descriptive and selective manner, and (3) it indirectly justified certain types of ‘discrimination’. Swendon and Windsor [
72] also mentioned that modern-day misunderstandings of multiculturalism in health care policy tended to perpetuate beliefs of ‘racial superiority’.
Professional ethics in light of restrictive healthcare policy
Managing services for undocumented migrants is affected by the laws and regulations in which a health facility is operating [
73]. It is clear that almost all international law and legal instruments have (theoretically) secured migrants human rights, including their right to health care [
74,
75]. However, there are diverse ways of interpreting the law when it comes to real practice. As a consequence, substantial country-to-country and within-country variations were observed. These variations concerned the types of migrants permitted to be insured, the type and range of services, and differing levels of financial protection [
73].
Dauvrin et al. [
31] displayed the variation in legal provisions for undocumented migrants in16 European countries according to the level of care by categorising the surveyed countries into 3 subsets, namely, (1) countries allowing migrants to enjoy (almost) the full range of care (eg, France, Italy, Spain, etc.), (2) countries only allowing access to emergency services and certain primary care services (eg, Austria, Belgium, Denmark, etc.) and (3) countries denying the right to access health services at almost all levels and types of care (eg, Finland and Sweden). Marrow [
76] raised a distinct case in San Francisco where illegal migrants were ‘semi-legalised’ by the recognition of their residence permit in the city, thereby, rights to care were endorsed to a larger extent in San Francisco than elsewhere in California or in many other states in the US.It should be noted that the legal instruments, which ratified migrant’s right to care, cannot be exercised perpetually; they have been influenced by, and have shifted with, political and economic changes. This phenomenon oftentimes led to bewilderment amongst healthcare management as some health practitioners have been unable to keep pace with the rapid changes. An intense debate was heard in 2012 in Spain when the government made a substantial change in the national health care system. This limited the right of non-Spanish inhabitants, who lacked legitimate residence permits, to access health services (except for maternal and emergency services). The reform was backed by the Court of Auditors which argued that insured Spanish citizens were bearing high healthcare costs incurred by non-Spanish citizens; thus, the reform was justifiable [
77].
Aside from the framework of civil law and regulation, the practices of healthcare providers were constructed under health professional norms and ethics, the primary intention of which is to secure the health interest of all human beings regardless of ethnicity or nationality [
47,
78].
It seems that ‘formal’ health professionals (eg., physicians and nurses) still have ‘margins’ or ‘loopholes’ which enable them to exercise their discretion in protecting the interests of patients, even though, to some extent, such practices contradict the law [
79]. Priebe et al. [
80] explained several strategies/tactics whereby physicians, who worked in cities where migrants’ right to care was restricted, circumnavigated the obstacles of limited entitlement to health benefits and avoided unwanted financial burdens on migrants. These tactics included referring their clients to charitable NGOs or ordering laboratory samples in the physicians’ name. Strabmayr et al. [
56] labelled such adaptive behaviour as ‘turning a blind eye’. Reporting the presence of illegal migrants to the police was undertaken only in special circumstances, such as when migrants were considering getting involving with crime or when they had risky behaviours which might pose a threat to the public [
58,
81].
In contrast to health professionals, supporting staff seemed to use those tactics less than health practitioners since the non-clinical staff were less bound by professional norms [
82]. Hargreaves et al. [
83] also observed that NHS payment officers in the UK, who were accountable for medical expense claims, had played a critical role in determining whether or not the overseas visitors were eligible to be exempted from charges for primary care service.
Vanthuyne et al. [
78] attempted to gain a deeper understanding of how providers strike a balance when torn between ‘professional ethics’ and ‘legal responsibilities’. Those arguing against universal access perceived illegal migrants to be abusing the host country’s health system and even expropriating resources (which were always sparse) from the native population; while on the other end of the continuum, some health professionals perceived uninsured migrants to be ‘deserving of free care’ on the basis of ‘right’. Interestingly, some respondents in that study found a compromise by designating migrants with precarious legal statuses as ‘vulnerable’ groups, whose ‘right to care’ became a ‘privilege’; thus care was given based on a principle of humanitarian aid or philanthropy, rather than as a ‘right’ [
78].
Interestingly, stricter and more complex rules governing the normalisation of migrants’ immigration status have not discouraged the influx of migrants. Though this was not specifically identified in any of the selected articles, this was demonstrated in some international publications. Having analysed the immigration history of Mexicans in the US, De Genova [
84] suggested that even in a period when immigration law became ostensibly stricter, it did not deter migration but rather generated a shift from legal to illegal migration. Van Der Leun [
85] pointed out that the Linking Act in the Netherlands, which aimed at excluding illegal migrants from using public services, caused an obvious tension on the local-level staff due to the way it shifted the responsibility for limiting migrants rights to care from immigration control officers at the country border, to local healthcare staff. The United Kingdom Trade and Investment Department suggested that the recent tightening of immigration laws in the UK would worsen the country’s current economic recession, and also increase the unemployment rate [
86].
The contradiction between the law and professional ethics might create tension and misunderstandings between healthcare providers and their migrant patients. Lyons et al. [
87] suggested that poor relationships between providers and patients might contribute to adverse effects on public health as a whole because migrants would be likely to sneak out from regular/formal health services, and therefore remain untreated. A relevant finding was presented by Biswas et al. [
5] who described how doctors in emergency wards in Denmark perceived a sense of mistrust amongst South Asian undocumented migrants, this was evidenced by the use of ‘false identification’ (using another person’s name when visiting a facility instead of using their real name) by some migrants when utilising services.
Strengths, weaknesses and limitations
This study has a key strength in gathering cutting edge evidence about how providers perceived, and adapted themselves in delivering care to migrants in their daily practice. However, despite a rigorously designed method, the review still had some weaknesses and limitations.
The first methodological limitation was that the search strategy did not encompass non-English-language articles due to limited interpreting capacity. Secondly, the majority of articles were retrieved from online databases and their selection was largely based on the MESH search strategy. Despite recruiting some grey literature from key international agencies, the grey literature from other sources, such as university-based reports, and unpublished articles and domestic text books, were likely to be left behind. This point is very important since migrants’ health is very context-specific. Individual country reports might have explored this topic more deeply than peer-reviewed publications.
Lastly, quality assessment was not executed in an enumerating/scoring system, which is conventionally done in most systematic reviews and meta-analysis. The reason for not using a quality assessment score stemmed from the fact that, since this review aimed to capture a broad understanding of the perceptions, attitudes and practices of healthcare workers providing services for immigrants, having a great miscellany of evidence (sensitivity) was deemed preferable than recruiting only studies with good quality (specificity). This is the so-called ‘configuring’ approach as described by Gough et al. [
88] and Voils et al. [
89]. The configuring approach is a method for synthesising research in which findings are used to explain and modify theoretical or narrative renderings of the target outcomes. Unlike configuration, ‘assimilation’ is an approach in which findings are pooled together in order to answer a specific research question [
88,
89]. Atkins et al. [
90] suggested that appraising the quality of qualitative studies might be an exercise in judging the quality of the written report rather than the research procedure per se. Articles published in qualitative-oriented journals were easier to evaluate since the length of articles allowed the authors to give details on the research process. Thus, evaluating the relative merits of the articles was not the primary concern in this case. In contrast, the quality reporting here aimed to remind the audience about the limitations of each study should its findings be applied in a real life setting.
Regarding limitations in the study results, firstly, it is important to remember that most of the selected studies were produced in developed countries in Europe; only two articles were from developing nations [
38,
58]. This issue might limit the generalisation of the review findings. It should be noted that there was no distinct difference in the study results between countries with differing economic statuses, or between countries with different health insurance systems. However, each country has introduced different rules and regulations in guaranteeing migrants’ right to health care (for example, the UK health system allows undocumented migrants to utilise emergency care, primary care, and treatment for some infectious diseases; in Switzerland, undocumented migrants must buy private health insurance under public supervision in the same way as Swiss citizens) [
91]. Future studies that deeply explore and collate evidence from countries with different types of health insurance models were recommended.
Secondly, the legal/citizenship status of migrants is very dynamic. Migrants with secured legal status may become illegal migrants if they stay in a host country longer than the visa permission; and, on the other hand, the status of undocumented migrants may be legalised once they register themselves with the state authorities. Most of the articles presented in this review focused on health services for migrants with precarious legal status, such as refugees, irregular migrants and undocumented persons. The review hardly explored the status of more affluent migrants, for instance, tourists, expatriates, and foreign businessmen. Accordingly, generalisations of the study’s findings to other types of migrants should be made with caution.
Thirdly, not all aspects of providers’ attitudes were explored. The review reported much about how healthcare providers addressed language barriers and contradictions between professional norms and the law, however, the measures taken to overcome challenges caused by different cultural and religious beliefs were sparsely reported.
Lastly, the reported perceptions and practices of healthcare providers demonstrated in this review were mainly drawn from the subjective assessment of the participants. Almost all of the articles employed in-depth interviews and focus group discussions as their primary data collection tools. Hence, it is possible that the reported perception of the respondents might be different from their real clinical practice. Further studies, which devise a variety of data collection techniques (such as observations, document studies, etc.) will be of great benefit in the development of appropriate healthcare system for migrants in the future.